Field of the Invention
The present invention relates to an image processing apparatus and an image processing method.
Description of the Related Art
An image printing apparatus is known which prints an image by repeatedly performing print scanning that discharges ink while moving a print head where a plurality of printing nozzles that discharge ink are arranged, relatively in a scanning direction with respect to a unit area of a print medium, and sub scanning that conveys the print medium in a conveyance direction being a direction intersecting the scanning direction. What is called a multipass printing method is known which forms an image by performing print scanning more than once on the unit area in such an image printing apparatus. In the known multipass printing method, image data having one-bit information that determines discharge or non-discharge of ink on a pixel by pixel basis is divided into a plurality of scans. The image data and a plurality of mask patterns having one-bit information that determines whether or not to allow ink to be discharged on a pixel by pixel basis, the plurality of mask patterns corresponding to a plurality of scans, are used to generate print data to be used for printing in the plurality of scans.
It is known to control an ink application order in accordance with the print condition when discharging a plurality of color inks in an image printing apparatus that performs a multipass printing method such as described above. Japanese Patent Application Laid-Open No. 2010-000666 discloses the use of a first mask pattern group including a plurality of mask patterns whose print permitting rates are substantially equal to each other, and a second mask pattern group including a plurality of mask patterns where the print permitting rates of mask patterns corresponding to scans of the second half are set to be higher than the print permitting rates of mask patterns corresponding to scans of the first half. According to Japanese Patent Application Laid-Open No. 2010-000666, if the amount of discharge of a cyan ink is larger than that of a magenta ink, the first mask pattern group is applied to image data corresponding to the cyan ink. Print data is generated such that the amounts of discharge are substantially equal to each other in a plurality of scans. Moreover, if the amount of discharge of the cyan ink is smaller than that of the magenta ink, the second mask pattern group is applied to image data corresponding to the cyan ink. Print data is generated such that the amount of discharge in the scans of the second half among the plurality of scans is larger than the amount of discharge in the scans of the first half. Consequently, it is possible not to especially specify the ink application order if the amount of discharge of the cyan ink is larger than that of the magenta ink, and to control the application order such that the cyan ink is applied onto a print medium later than the magenta ink if the amount of discharge of the cyan ink is smaller than that of the magenta ink.
On the other hand, in recent years, it has been known to generate print data using image data having multiple-bit information that determines the number of discharges of ink on a pixel by pixel basis, and a plurality of mask patterns having multiple-bit information that determines the allowable number of discharges of ink on a pixel by pixel basis, the plurality of mask patterns corresponding to a plurality of scans. Print data is generated in this manner to make it possible to discharge ink more than once to one pixel area. For example, a case where pieces of information respectively configuring image data and a mask pattern are two bits is described in Japanese Patent Application Laid-Open No. 2003-175592.
However, if image data and a mask pattern, which each include multiple-bit information, are used to discharge ink more than once to one pixel area, the number of bits of information configuring the image data and the mask pattern is increased. Consequently, memory with a certain degree of large capacity may be required as compared to the known case where image data and a mask pattern, which each include one-bit information per pixel, are used.
It may be required here to store a plurality of mask pattern groups according to the print conditions for image data corresponding to a given ink in order to control the ink application order according to the print condition if the above-mentioned technology described in Japanese Patent Application Laid-Open No. 2010-000666 is used. In other words, a first and a second mask pattern groups different from each other are stored in advance. If the print condition is a first print condition, the first mask pattern group is applied to generate print data. If the print condition is a second print condition that is different from the first print condition, the second mask pattern group is applied to generate print data. A plurality of mask pattern groups each having multiple-bit information per pixel is stored in this manner, and accordingly larger-capacity memory may be required.
Furthermore, a mask pattern is generally designed at a certain degree of size (for example, a size of 12 pixels in total: six pixels in an X direction and two pixels in a Y direction). In the technology described in Japanese Patent Application Laid-Open No. 2010-000666, the mask pattern group to be applied is changed according to the print condition. Accordingly, the ink application order can be controlled only by size of the mask pattern. A size smaller than the size of the mask pattern is assumed to be a size for determining a print condition (for example, a size of six pixels in total: three pixels in the X direction and two pixels in the Y direction). A mask pattern is divided according to the determination size to be applied. Accordingly, the application order can be controlled by size that is smaller than the size of the mask pattern. However, the processing load of a control unit is increased.